The maximality conjecture for Gaussian Riemann derivatives

Let nn be a positive integer, and let DAD_{\mathcal{A}} be an nn-th generalized Riemann differentiation without excess. A generalized Riemann differentiation is called Gaussian when it is one of the Gaussian Riemann derivatives introduced in the paper, and symmetric Gaussian when it is the corresponding symmetric derivative. For a function ff and point xx, write f(k)(x)f_{(k)}(x) and f(k)s(x)f_{(k)}^s(x) for the kk-th Peano and symmetric Peano derivatives, respectively.

Maximality conjecture for Gaussian Riemann derivatives. For all functions ff and points xx:

  1. If DAD_{\mathcal{A}} is not Gaussian, then
both f(n1)(x) and DAf(x) exist  f(n)(x) exists\text{both }f_{(n-1)}(x)\text{ and }D_{\mathcal{A}}f(x)\text{ exist}\ \Longrightarrow\ f_{(n)}(x)\text{ exists}

should be false.

  1. If n3n\geq 3 and DAD_{\mathcal{A}} is symmetric but not symmetric Gaussian, then
both f(n2)s(x) and DAf(x) exist  f(n)s(x) exists\text{both }f_{(n-2)}^s(x)\text{ and }D_{\mathcal{A}}f(x)\text{ exist}\ \Longrightarrow\ f_{(n)}^s(x)\text{ exists}

should be false.

The conjecture identifies the Gaussian and symmetric Gaussian derivatives as the largest classes for which the generalized Marcinkiewicz–Zygmund equivalences can hold. Part (i) is known for n=1,2n=1,2 and remains open for n3n\geq 3; part (ii) is false for n=1,2n=1,2, true for n=3,4n=3,4, and remains open for n5n\geq 5.

Sources & referencesView supporting material

Primary source

J. M. Ash, S. Catoiu and H. Fejzic, “Gaussian Riemann Derivatives”, arXiv:2211.09209 (2022).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.